Date of Publication :7th April 2016

Abstract: Information transmitted in the form of digital images is important part of communication system, but restored image obtained after transmission is often corrupted with noise. The restored image needs processing before it can be used in applications. Aim of the image processing is to visually enhance some aspect of an image which is not readily apparent in its original form. Image processing gives greater perception & vision but it does not add any information content. Image captured by any system is the degraded version of an original image. The degradation is due to sensing environment when it acquire through optical, electro optical and electronics medium. It is necessary to restore such image for further image processing & other tasks. By keeping the integrity of the image information, Gaussian noise removal is a critical issue. The function of the proposed algorithm is to replace each corrupted pixel is by alpha trimmed value of pixels inside an adaptive window. Using threshold which is calculated from noise variance the adaptive window is formed. According to Peak signal to noise ratio (PSNR) performance proposed algorithm is better and effective for images which are highly corrupted by Gaussian noise

Reference :